[ADMB Users] Accounting for autocorrelation in glmmADMB using R

H. Skaug hskaug at gmail.com
Wed Apr 13 00:36:39 PDT 2011


Hi,

It is not possible within glmmADMB, but if you are willing to start using ADMB
then it is easy. I attach an AR1 model with negative binomial response.
(I agree that the code may look confusing, but my point is that there
are only a few lines of code that does the job so it cannot possibly
be that difficult :-)
Here are pointers to spatial models:

http://admb-project.org/examples/spatial-models/glmm2019s-on-large-spatial-grids
http://admb-project.org/examples/spatial-models/glmm-with-spatial-structure-described-in-terms-of-covariance-function

By working directly with ADMB you will be able to get more help from this list.
Also there exist packages for calling ADMB programs from R.

Hans


On Tue, Apr 12, 2011 at 8:08 PM, Holly G <hollygoy at hotmail.com> wrote:
> Hi,
> Do any users of glmmADMB in R recommend a straight-forward way of accounting
> for autocorrelation (spatial or AR1 structure)?
> Thank you,
> Holly
>
> _______________________________________________
> Users mailing list
> Users at admb-project.org
> http://lists.admb-project.org/mailman/listinfo/users
>
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: polio_nb.tpl
Type: application/octet-stream
Size: 2309 bytes
Desc: not available
URL: <http://lists.admb-project.org/pipermail/users/attachments/20110413/067ffc83/attachment.obj>


More information about the Users mailing list